Multi-site analysis of the impact of surface ozone on European forests

Tropospheric Ozone (O₃) is a secondary pollutant with a positive radiative forcing and many negative effects on air quality, human health and ecosystems at different scales. In fact, O₃ acts as a strong oxidant in plants, negatively impacting many cellular and molecular processes, such as modifying rubisco activity, reducing stomatal conductance and inducing early leaf senescence. Furthermore, when the O₃ levels at the surface are high (typically above 40 ppb), these combined effects may decrease the photosynthetic carbon gains, mainly detectable at the leaf level but also important at the tree and stand scales. In this study, we intend to evaluate the effects of O₃ on Gross Primary Production (GPP) at four European forest sites (Belgium, France and Italy) using local measurements, both for O₃, eddy-covariance (GPP) and meteorological variables (including air temperature (TA), relative humidity (RH), vapour pressure deficit (VPD), short-wave radiation (SW), wind and precipitation). We first applied a series of statistical analyses to identify the impact of O₃ and meteorological variables on GPP for each site. In a second step, we used a process-based model that simulates GPP using the Farquhar equations parameterised for each site to quantify O₃-induced GPP reductions. Our results showed that the dominant meteo factor is site dependent. The SW appears to be the most important variable for predicting GPP at all sites, contrary to TA and VPD. The Belgium and Italian sites show a correlation of 0.51(0.55) for TA and 0.38(0.48) for VPD, respectively. Whereas in the French site, the TA correlation of 0.55 exceeds the VPD correlation of 0.38. Consequently, the GPP reduction varies along the different sites. This study shows the necessity of long-term monitoring datasets to understand better the O₃ impacts at several ecosystems combined with process-based models.